Grant: Doctoral Research Award: Canada Graduate Scholarships
McMaster University
$120,000.00
AI for exploring ultra-large chemical space for novel antibiotic scaffolds with readily accessible analogs for easy hit-to-lead drug development.
FY 2024-25
Summary
Antimicrobial resistance (AMR) poses a critical threat to global health, with predictions suggesting it could lead to millions of deaths in the coming decades. One major challenge contributing to this crisis is the lack of new antibiotic classes in recent decades. While artificial intelligence has shown promise in exploring vast virtual chemical libraries (>30 billion) for new antibacterial molecules, early promising candidates often fail because of their sensitivity to change. Chemical 'edits' to their structure can disrupt their antibacterial properties, meaning unless the specific compound already satisfies dozens of properties, such as metabolic profile and toxicity, it will lead to a dead end. This highlights the need to enhance the initial screening process to include predictions of how easily modifications can be made while preserving effectiveness against bacteria.
This project aims to develop an algorithm that identifies "predictophores", chemical "cores" integral to a molecule's predicted antibacterial activity. These predictophores will be crucial for quickly designing and optimizing new antibacterial compounds. The study focuses on Staphylococcus aureus, a significant pathogen, using Enamine's Readily Accessible (REAL) space, which contains over 30 billion molecules designed for efficient synthesis and low cost.
The project's aims to 1) Developing an algorithm to efficiently search large chemical databases for predictophores. 2) Validating these predictophores by testing synthesized molecules for their ability to inhibit S. aureus growth in laboratory settings. 3) Optimizing promising molecules for multiple clinical properties such as human toxicity and solubility while maintaining their antibacterial activity.
Successful outcomes could lead to the development of new classes of antibiotics that are both effective against resistant bacteria and adaptable to various clinical needs, potentially transforming the landscape of antibiotic discovery.
Department
Program
Doctoral Research Award: Canada Graduate Scholarships
Recipient
Location
Ontario, Canada
Recipient Type
Academia
Type
Grant
Source
CIHR Funding Decisions Database
As published
Id
518233
Name
Liu Gary
Region
Ontario
Country
Canada
Orgtype
Institution Paid; Research Institution
Instname2
Infection and Immunity
Theme2
Biomedical
Prcname2
Doctoral Research Awards - A
Projecttitle
AI for exploring ultra-large chemical space for novel antibiotic scaffolds with readily accessible analogs for easy hit-to-lead drug development.
Programtype2
Studentship Programs
Orgnameinp2
McMaster University
Cihramount2
$120,000
Cihrequipment2
$0
Programname2
Doctoral Research Award: Canada Graduate Scholarships
Keyworddelim
Antibiotics; Applied Statistics; Biochemistry; Chemistry; Computational; Computational Chemistry; Deep Learning; Drug Discovery; Interdiciplinary; Machine Learning
Abstract
Antimicrobial resistance (AMR) poses a critical threat to global health, with predictions suggesting it could lead to millions of deaths in the coming decades. One major challenge contributing to this crisis is the lack of new antibiotic classes in recent decades. While artificial intelligence has shown promise in exploring vast virtual chemical libraries (>30 billion) for new antibacterial molecules, early promising candidates often fail because of their sensitivity to change. Chemical 'edits' to their structure can disrupt their antibacterial properties, meaning unless the specific compound already satisfies dozens of properties, such as metabolic profile and toxicity, it will lead to a dead end. This highlights the need to enhance the initial screening process to include predictions of how easily modifications can be made while preserving effectiveness against bacteria.
This project aims to develop an algorithm that identifies "predictophores", chemical "cores" integral to a molecule's predicted antibacterial activity. These predictophores will be crucial for quickly designing and optimizing new antibacterial compounds. The study focuses on Staphylococcus aureus, a significant pathogen, using Enamine's Readily Accessible (REAL) space, which contains over 30 billion molecules designed for efficient synthesis and low cost.
The project's aims to 1) Developing an algorithm to efficiently search large chemical databases for predictophores. 2) Validating these predictophores by testing synthesized molecules for their ability to inhibit S. aureus growth in laboratory settings. 3) Optimizing promising molecules for multiple clinical properties such as human toxicity and solubility while maintaining their antibacterial activity.
Successful outcomes could lead to the development of new classes of antibiotics that are both effective against resistant bacteria and adaptable to various clinical needs, potentially transforming the landscape of antibiotic discovery.
Competitiondate
202411
Approvedterm2
3 yrs 0 mth
Pinamesdelim
Liu, Gary
Approvedterm
3 ans 0 mois
Cihrcontribution2
~ $120,000~ $0~